THE LAW OF JOHANNSEN. 215 



between alloys can be due to relative preponderance of one or 

 several constituent metals. And therefore, such differences be- 

 tween cells can be thought of as ultimately due in a great mea- 

 sure to the "environment" of the cells in as far as it furnishes 

 the "ingredients" for some genes in excess. If we conceive 

 of the genes as of chemical substances with autokatalytic prop- 

 erties, rather than as of "vital" protoplasmic "determinants," 

 we might take as a simple illustration any relatively little 

 complex autokatalytic substance, such as oxide of iron, and 

 imagine it for a moment to be a gene and a factor in the devel 

 opment of some organism. We will see how a few generations 

 of cells in an evironment with plenty of iron and oxygen, have 

 come to contain a great quantity of this gene, Fe0 2 . The pres- 

 ence of much of this gene may well become appreciable as a 

 colour, let us say red. Now, even if the cause of this redness 

 changes, if the circumstances which brought the cells into con- 

 tact with iron and oxygen, alter the daughter-cells of a red 

 mother-cell will be redder than the average. No matter how 

 abrupt this change will be, the change in characters, colour 

 through cell-generations will be comparatively gradual. 



If, by some uncommon constitution of the environment, a sea- 

 urchin egg shows very dark colour, its two, its four blastomeres, 

 will yet be redder, than the common run of cells in twocell and 

 four-cell stages. We might measure the coefficient of correlation 

 and f indit very high between a mother-ceil andits daughter-cells . 



Now, if we remember that those daughter-cells a moment ago 

 were the mother-cell, we begin to ask ourselves the question : 

 How does this resemblance between mother-cell and daughter- 

 cell, between egg and blastomeres in a sea-urchin compare to 

 the resemblance between a mother-sea-urchin and a daughter- 

 sea-urchin ? 



The process of transition between a mother-cell and its two 

 daughter-cells is so direct, that all the factors in the develop- 

 ment of the mother-cell, genes as well as non-genetic factors, 

 are also factors in that of the daughter-cells. The quality of a 

 cell is the result of its genotype, and of the way in which it has 



